Basic Emotion Questions Er.Sagepub.Com

Home , Emotion work, Surprise (emotion)

Emotion Review Vol. 3, No. 4 (October 2011) 379–386 © The Author(s) 2011 ISSN 1754-0739 DOI: 10.1177/1754073911410743 Basic Emotion Questions er.sagepub.com

Robert W. Levenson Department of Psychology, University of California, Berkeley, USA

Abstract

Among discrete emotions, basic emotions are the most elemental; most distinct; most continuous across species, time, and place; and most intimately related to survival-critical functions. For an emotion to be afforded basic emotion status it must meet criteria of: (a) distinctness (primarily in behavioral and physiological characteristics), (b) hard-wiredness (circuitry built into the nervous system), and (c) functionality (provides a generalized solution to a particular survival-relevant challenge or opportunity). A set of six emotions that most clearly meet these criteria (enjoyment, anger, disgust, fear, surprise, sadness) and three additional emotions (relief/contentment, interest, love) for which the evidence is not yet quite as strong is described. Empirical approaches that are most and least useful for establishing basic-emotion status are discussed. Basic emotions are thought to have a central organizing mechanism and to have the capacity to influence behavior, thoughts, and other fundamental processes.

Keywords basic emotions, discrete emotions, emotional language, evolutionary approaches

Question 1: How does a basic emotion differ from simply a Only a subset of discrete emotions is basic. Based on my criteria discrete emotion? those basic emotions must have qualities of distinctness, hard- wiredness, and functionality; I believe that the best evidence Basic emotions are a special class of discrete emotions, a sub- exists for six basic emotions (with associated functions in set of emotions that are most elemental; most distinct; most parentheses): (a) enjoyment (playing), (b) anger (fighting), continuous across species, time, and place; and most intimately (c) disgust (rejecting), (d) fear (avoiding), (e) surprise (orient- related to survival-critical functions. Because basic emotions ing), and (f) sadness (help seeking). There are three additional are a subgroup of discrete emotions, they inherit key features emotions that I believe are basic, with well-established func- shared by all discrete emotions. These include separateness tionality, but for which the existing evidence for distinctness (there are clear boundaries that distinguish one discrete emo- and/or hard-wiredness is not as strong: (g) relief/contentment tion from another), finiteness (the number of discrete emotions (soothing), (h) interest (exploration), and (i) love (attachment). is limited), and singularity (the recipe for a discrete emotion Although mostly single emotion terms are used in this list of cannot be reproduced by blending varying quantities of dimen- basic emotions, in reality each represents a “family” (Rosch & sional ingredients such as valence and arousal). Lloyd, 1978; Rosch & Mervis, 1975; Russell, 1991) of closely Theorists differ as to the particular characteristics that are related emotions. Thus, the basic-emotion family of sadness considered necessary and sufficient for affording basic emotion would include emotions such as distress and anguish. status to a discrete emotion. In my view, these characteristics are: Distinguishing basic emotions from other emotions requires (a) distinctness (primarily in behavioral and physiological char- proposing criteria, postulating the evidentiary basis for each acteristics), (b) hard-wiredness (circuitry built into the nervous criterion, and evaluating the state of the existing evidence. Any system), and (c) functionality (provides a generalized solution to discussion of criteria for basic emotions must start with a particular survival-relevant challenge or opportunity). Ekman’s (1992) nine criteria. With some aggregation, his list Question 2: What is your list of basic emotions? Are all can be reduced into three larger groupings: (a) distinctness (in emotions basic, or just some? If some, how do you distin- antecedents, signal, and physiology), (b) continuity (presence in guish basic from nonbasic emotions? What is the relation other primate species, same antecedents in all cultures, same of nonbasic to basic emotions? signals in all cultures), and (c) structure/function (quick onset,

Author note: Preparation of this manuscript was supported by Grants AG17766 and AG19724. Corresponding author: Robert W. Levenson, Department of Psychology, University of California, 3210 Tolman Hall #1650, Berkeley, CA 94720-1650, USA. Email: [email protected] 380 Emotion Review Vol. 3 No. 4

brief, unbidden, coherence among responses, automatic appraisal). for illuminating the brain regions and circuits that elicit and My three criteria borrow from these larger groupings, including disrupt particular emotions. distinctness (but primarily established by evidence of modular- ity in the nervous system), hard-wiredness (inferred from evidence of continuity across species, development, and place), and Hard-Wiredness functionality (with greater concern for antecedent conditions and less concern for particular structural qualities). Establishing hard-wiredness requires that the circuitry for a basic emotion is built into the nervous system in at least some primitive form. Learning and experience may shape and Distinctness enhance the original circuitry, but if an emotion has to be learned de novo, then that emotion is not basic. Establishing distinctness requires a basic emotion to have When attempting to establish hard-wiredness, researchers qualities that distinguish it from other emotions (in particular, have most often sought evidence of continuity across species, from other basic emotions). Traditionally, evidence for distinct- development, and cultures. Each of these approaches has ness has taken the form of differences in behavior (e.g., motor value, but also has limits. First, each approach must be accom- action patterns), expression (e.g., facial and vocal signals), panied by the caveat that evidence of continuity is being physiology (e.g., autonomic or central nervous system activity), extrapolated from a particular subsample of species, develop- and, if appropriate to the species, language (e.g., emotion mental stages, and cultures and, thus, does not establish words, self-reports of emotional experience). I believe that universality for all species, stages, and cultures. Second, these approaches remain largely valid, but will argue later judgments as to what constitutes similarity are complicated by (Question 8) that, even when language is available, behavior ontogenetic and phylogenetic differences among species and and physiology should be afforded greater weight. developmental stages, and by language differences among Alternative approaches to establishing distinctness, which cultures. Third, using cross-cultural studies (and even more so may ultimately prove to be even more useful, make use of neu- for cross-national studies) to establish universality has become rological patients and stimulation/blocking of particular brain particularly problematic. With any demonstration of similari- circuits. With these methods, support for distinctness is obtained ties between disparate cultures, there is an alternative explana- when particular emotions appear or disappear in particular tion that the cultures, facing similar kinds of problems and neurological diseases or when particular brain circuits are possessing similar resources, could have learned the same stimulated or blocked. emotion in the same way. Researchers have traditionally As an example of the utility of using neurological patients sought to maximize isolation and maximize differences between for exploring distinctness, we (Olney et al., 2011) have recently cultures as a way of minimizing this possibility. However, this been working with a group of patients with amyotrophic lateral approach can never totally eliminate the alternative learning sclerosis (a progressive motor neuron disease). Many of these explanation. Moreover, globalization and ecological changes patients manifest pseudobulbar affect (Wilson, 1924), which have made cultural isolation essentially nonexistent. Thus, consists of spontaneous episodes of uncontrolled laughing and emotional similarities that are found between even the most crying. These episodes include intense “prototypical” facial seemingly disparate cultures no longer have the cachet they expressions of happiness and sadness, coupled with high- once had for establishing hard-wiring. magnitude autonomic and respiratory responses consistent with these emotions. Some of these patients laugh, some cry, and some do both. But we have never seen any instances of sponta- Functionality neous uncontrolled anger, fear, disgust, or surprise, supporting distinctness between the emotions that occur spontaneously and Establishing functionality requires evidence that a basic emotion those that do not. A related approach, focusing on emotional addresses a particular challenge or opportunity or solves a par- deficits, is seen in studies of patients with Huntington’s disease. ticular problem that is critical to species survival and thriving. These patients have been reported to have deficits in the Often this particularity is established by well-reasoned asser- processing and production of disgust but not in other emotions tion. Thus, anger is said to be the solution for the problem of (Gray, Young, Barker, Curtis, & Gibson, 1997; Sprengelmeyer, needing to keep control of that which is ours; a functional Young, Calder, & Karnat, 1996; Sprengelmeyer et al., 1997). explanation that seems reasonable and compelling. The force of Examples of studies using stimulation and blockade method- these assertions can be strengthened by careful consideration of ologies have also been informative. In cats, the elicitation of a the evolutionary environments of our ancestors, the likely prob- particular emotion, anger (so-called “sham rage”), has been lems they encountered, and the likely resources they possessed associated with acute brainstem transection and amygdala (Tooby & Cosmides, 1990). stimulation (Reis & Gunne, 1965; Reis, Miura, Weinbren, & In the more empirical realm, functionality is often estab- Gunne, 1967). Studies using transcranial magnetic stimulation lished by searching for particular antecedent conditions that (George et al., 1996) and selective pharmacological blockade are associated with particular emotions or that elicit these (Lawrence, Calder, McGowan, & Grasby, 2002) show promise emotions. A number of studies have sought evidence of Levenson Basic Emotion Questions 381

cross-cultural consistencies in these antecedent conditions For an emotion to be basic, there needs to be a central organizing (Boucher, 1983; Matsumoto, Kudoh, Scherer, & Wallbott, mechanism. In my view, this mechanism operates by searching 1988; Scherer, Summerfield, & Wallbott, 1983). Another continuously for meaningful patterns in incoming sensory infor- approach has been to examine the consequences of particular mation, recognizing survival-critical situations, and activating emotions. For example, we (Fredrickson & Levenson, 1998; the appropriate emotion, which recruits and orchestrates the Yuan, McCarthy, Holley, & Levenson, 2010), have demon- optimal behavioral and physiological responses. In an earlier strated that efficient reduction of physiological arousal pro- formulation (Levenson, 1999), I described this central organiz- duced by negative emotions is a function of emotions such as ing mechanism as the “core system,” recognizing its debt to enjoyment and contentment (but not of other basic emotions). Tomkins’ “affect program” (Tomkins, 1962). In my view: Similarly, in building their case for embarrassment as a basic emotion, Keltner and Buswell (1997) accumulated extensive the core system has all of the capabilities necessary for processing evidence that its primary function is appeasement. incoming information continuously and for detecting a small number of prototypical situations that have profound implications for the organism’s immediate well-being and long-term survival. Having recognized Controversies in the stream of incoming perceptual information the configuration of features that defines one of a small number of prototypical situations, the core system activates an emotion, which is comprised of a set of Major disagreements have arisen in the basic-emotion literature response tendencies that have been selected by evolution for their high over the methodologies used to collect data and the evidentiary probability of dealing successfully and efficiently with the problems thresholds for meeting each criterion. A classic example can be posed by that particular situation. The configural features of the proto- found in the study of facial expressions associated with different typical situation and the exact features of the response package that is emotions, one of the most extensively studied aspects of emotion recruited differ from species to species (e.g., differences across species in the literatures related to both distinctness and hard-wiredness. in what constitutes a predator and in the acoustic qualities of a fear vocalization), but the basic purpose and operation of the core system is In terms of methodology, most studies have had participants the same (i.e., matching environmental events to prototypes, recruiting identify the emotions depicted in photographs of emotional and orchestrating the appropriate response). (Levenson, 1999, p. 484) expressions. But, these studies can all be criticized for using a methodology based on emotion recognition and not on emotion Thus, my view of a basic emotion’s organizing mechanism production. In terms of evidentiary thresholds, almost all of the includes: (a) a very fast, low-level pattern detector on the input “classic” studies found agreement across cultures at greater than side tuned to survival-significant events, and (b) a very fast, low- chance levels for recognizing emotions such as anger, contempt, level mechanism on the output side that efficiently recruits and disgust, fear, happiness, sadness, and surprise (Ekman & Friesen, organizes disparate perceptual, behavioral, and physiological 1971; Ekman & Heider, 1988; Izard, 1971). Moreover, a newer systems to produce a response most likely to provide a successful generation of these studies has argued that cross-cultural agree- solution to the challenges or opportunities posed by these events. ment also exists for self-conscious emotions such as embarrass- Extrapolating from my answer to Question 2, the central ment and pride (Keltner, 1995; Tracy & Robins, 2008), and some organizing mechanism would have to be initially hard-wired in recent studies have examined production as well (Elfenbein, the nervous system in at least a primitive form. However, the Beaupre, Levesque, & Hess, 2007; Matsumoto & Willingham, control mechanisms that regulate it can be greatly influenced by 2006, 2009; Tracy & Matsumoto, 2008). However, in all of these learning, experience, and, in humans, higher order processes studies, both old and new, the level of agreement varies across such as goals, values, and social norms. Previously (Levenson, cultures and emotions, and perfect agreement has been rare. This 1999) I described the interplay between the central organizing begs the question of how much agreement is “enough.” and control mechanisms as follows: Finally, when multiple criteria are proposed, there can be disa- greement over how many must be met: one, some, or all? I have At the core of the emotion system is a remarkably durable, simple, and posited three criteria that I believe are essential for affording efficient “processor”, designed early in evolution to cope effectively basic-emotion status: distinctness, hard-wiredness, and function- with a few very basic, ubiquitous problems (e.g., Ekman, 1992; Lazarus, ality. For an emotion to be deemed basic, evidence in support of 1991; Levenson, 1994; Tooby & Cosmides, 1990) in time-tested, highly predictable, and quite automatic (e.g., Zajonc, 1984) ways. Surrounding all three criteria would be necessary and sufficient. However, this this core system is a more recently evolved, highly flexible, and much is not a case of one size fitting all. For example, evidence of dis- less predictable set of control mechanisms that are designed to influence tinctness for one basic emotion might derive from evidence of the actions of the core system. Whereas the core system is largely hard- universality of facial expression, and for another from evidence of wired and not capable of major modification in response to experience, unique blocking by transcranial stimulation. Moreover, confi- the control mechanisms are exquisitely sensitive to learning, fine-tuning dence that an emotion is basic will surely increase to the extent their operating parameters across the course of life […] The control that multiple sources of evidence exist in support of each criterion. system acts on the “input” to the core system by altering the conditions that set the core system into action (e.g., Lazarus, 1991; Scherer, 1984; Smith & Ellsworth, 1985), and it acts on the “output” of the core system Question 3: Does the existence of a basic emotion depend on by intercepting tendencies to respond to prototypic situations in charac- the existence of a central organizing mechanism (something teristic, stereotypical ways and modulating the translation between like an “affect program”) or can a basic emotion be simply response tendencies and resultant behavior (e.g., Ekman & Friesen, a patterned response? 1969; Hochschild, 1979). (Levenson, 1999, pp. 483–484) 382 Emotion Review Vol. 3 No. 4

In my view, the central organizing mechanism retains the In my view, a basic emotion has to be evolutionarily shaped and capacity to override these control mechanisms and other higher biologically prewired. Basic emotions arise from the need for order cognitive processes under conditions where challenges solutions to a small set of prototypical challenges, problems, and opportunities appear clearly, suddenly, and intensely. One and opportunities that have profound implications for species important implication of this for studies of basic emotions is survival and thriving. In mammalian species, these include such that their organizing effects will be seen most clearly when things as bonding with others, handling threats, dealing with elicitors are focused, powerful, sudden, and closely match pro- loss, defending what is ours, avoiding noxious substances, and totypical antecedent conditions (Levenson, Soto, & Pole, soothing self and others. The ubiquity and importance of these 2007). In social science experiments, this is often not the case. challenges, problems, and opportunities would have created Rather, in these studies, emotional stimuli are typically mild, enormous selection pressures favoring generalized solutions for gradual in onset, diffuse, and do not closely match prototypes. each—solutions that have the highest likelihood of producing Under these latter conditions, the observed response is likely to beneficial outcomes for the individual and for the group most of be a complex mixture of responses associated with the basic the time. Thus, basic emotions are the time-tested solutions to emotion combined with the responses associated with emotion these timeless problems. Because of their importance, having regulatory (Gross & Levenson, 1997) and other higher order each individual learn each solution de novo would be inefficient processes. and uncertain. For this reason, these well-honed solutions are best situated in the original hard-wiring of the nervous system. Question 4: In everyday discourse, emotions cause certain Beyond this, basic emotions differ in many ways. Some, like behaviors (fear makes us flee, makes our heart race, makes fear, appear earlier in phylogeny and ontogeny, have relatively us think irrationally, and so on). In your theory, does a simple neural circuitry, and thus could be said to be psychologi- basic emotion have such causal powers? Which powers? cally primitive. However, this seems more relative than absolute. For this reason, I would be reluctant to stipulate psychological Basic emotions certainly have the power to influence behaviors, primitiveness as a requirement for basic-emotion status. thoughts, and many other fundamental processes. However, Similarly, basic emotions may well serve as building blocks rather than use the term “cause,” I view this influence as more for other more complex emotions and for more complex emotion- probabilistic. Thus, in an earlier theoretical formulation related states. However, this is more an example of nature (Levenson, 1994a), I stated that emotions “alter attention, shift sometimes reusing its good ideas rather than a requirement. certain behaviors upward in response hierarchies, and activate Moreover, I do not endorse the view that complex emotions are relevant associative networks in memory” (p. 124). I think the all composed from a single palette comprised by the basic emo- terms “alter,” “shift,” and “activate” in this statement better tions. To reduce emotions such as pride, shame, embarrassment, capture what emotions do than does the term “cause.” In that awe, or guilt into X units of one basic emotion and Y units of earlier formulation I gave some examples of this process: another makes little sense. Although there may be some happiness that goes into pride and some sadness that goes into shame, Emotions have the capacity to activate certain behaviors, which might each also has emergent qualities that cannot be explained in normally exist at the bottom of behavioral hierarchies. Thus, under the terms of the small set of emotions I consider to be basic. proper conditions, anger can drive the pacifist to fight; sadness can make the strong weep; and fear can cause the brave to cower. In this Question 6: How are basic emotions differentiated one from regard, emotion has the unique capacity to set aside, in a moment, a lifetime of individualized learning, refinement, culture, and style, another? revealing the common denominator of human response. (Levenson, 1994a, p. 124) Basic emotions may differ from each other in numerous ways, but the essential differentiation begins with function. If you Combining this notion with my views about the conditions accept that basic emotions evolved to help to solve species- under which basic emotions have the capacity to override specific problems (Tooby & Cosmides, 1990) in highly effi- regulatory and higher processes (Question 3), I believe that the cient, generalized ways (Levenson, 1999) then, for problems influence of basic emotions on behaviors and thoughts becomes with a hard-wired solution, that solution should be embodied in most deterministic under those conditions in which antecedent a single discrete emotion. From this functionalistic first princi- conditions closely match prototypical elicitors, eliciting ple, other differentiating features arise. Thus, in social species conditions onset most suddenly, and emotions are most intense there are great advantages when an individual is in the throes of (Levenson et al., 2007). When these conditions are not met, the a basic emotion for that emotional state to be communicated to plasticity and flexibility of the emotion system becomes more conspecifics. The origins of the morphology of these signals has ascendant. long been a topic of great interest (Darwin, 1872). In my view, most of these signals derive from the behavioral and physiolog- Question 5: In what sense are basic emotions basic? ical adjustments the basic emotion orchestrates in fine-tuning Specifically, please touch on the questions about what makes the organism to best be able to deal with the problem at hand. a basic emotion basic: must the emotion be evolutionarily Depending on the response system, the specific indicator, and shaped? Biologically prewired? Psychologically primitive? A the basic emotions being considered, the differentiation of building block of other emotions? All of the above? signals by emotion can range from partial to complete. Levenson Basic Emotion Questions 383

Facial expressions are informative in this regard. Question 7: If your list of basic emotions is a set of English Functionalist views typically begin with speculation as to the terms, how do you respond to the claim that some languages purpose that might have originally been served by a particular lack equivalent terms for those emotions but include emotion muscle contraction or set of contractions (for a supportive terms that differ in meaning from English terms? What is analysis, see Susskind & Anderson, 2008). Thus, raising the the relation between your basic emotions and the everyday eyelid (AU5 in the Facial Action Coding System; Ekman & folk language people use to talk about their emotions? Friesen, 1978) could serve to increase the amount of light entering the eye, improving visual acuity. This kind of fine- In my view, the ways that we think about, label, and talk about tuning of the visual system would be useful in dealing with a our emotions are socially constructed; they are largely deter- number of emotion-eliciting situations. Following this logic mined by learning, social influence, and culture (Levenson et al., further, it is not surprising to find that AU5 appears in the pro- 2007). Although partially grounded in actual emotional totypical facial expressions (Ekman & Friesen, 1975) for a phenomena, emotion language is highly influenced by customs, number of basic emotions, including surprise, fear, and anger. mores, traditions, self-presentation biases, and cultural values. In all three emotions, the utility of an adjustment in the facial Emotion language may provide important clues about underly- muscles that increases visual acuity is apparent. However, AU5 ing emotional phenomena, but it cannot be considered defini- is not sufficient for identifying any one particular basic emo- tive in adjudicating basic-emotion status according to my three tion. Rather, it provides a useful clue for distinguishing one criteria of discreteness, hard-wiredness, and functionality. Thus, subset of basic emotions (fear, anger, surprise) from another a particular culture may not have a term for “sadness.” subset of basic emotions that do not include this action (e.g., However, this does not mean that members of that culture do enjoyment, sadness, and disgust). Another action unit, AU9, not possess the neural circuitry for a distinct sadness response which produces visible wrinkling of skin along the sides of the that mobilizes resources for dealing with loss. Similarly, a nose, is thought to serve the function of closing off the airways culture may have a term for an emotion unlike that found in to noxious odors. Unlike AU5, which is found in several basic any other culture. The existence of that term does not mean emotions, AU9 is only found in one, appearing in disgust but that members of that culture possess discrete, hard-wired, not in other basic emotions such as enjoyment, anger, fear, functional circuitry for that emotion. sadness, or surprise. Emotion language may not be definitive in affording basic- Autonomic nervous system adjustments also assume an emotion status, but it is still enormously important. Recent theo- important position in most functional views of emotion, acting retical and empirical work has explored the ways that cultures to produce the optimal bodily milieu for supporting the behavioral differ in their values and beliefs about what emotions one should adaptations associated with basic emotions (Levenson, 1988, have (Tsai, Knutson, & Fung, 2006), extending earlier work on 1992, 1994b). In addition to producing measurable changes in cultural feeling rules (Hochschild, 1979) and display rules bodily functions (e.g., alterations in heart rate, skin conductance, (Ekman & Friesen, 1969; Friesen, 1972). Nonetheless, I believe etc.), many autonomic nervous system adjustments produce these cultural values and beliefs have greater influence on the appearance changes that are highly visible (Levenson, 2003) and ways that people talk about, think about, and label their emo- have signal value that rivals that of changes in facial expression tional experiences than on the more behavioral and physiologi- (e.g., changes in coloration, piloerection). As with particular cal aspects of emotion (Levenson, Ekman, Heider, & Friesen, facial actions, a particular autonomic change may distinguish 1992; Levenson et al., 2007; Soto, Levenson, & Ebling, 2005). among subsets of basic emotions rather than being unique to a Emotion language can also shed light on issues of funda- particular emotion. Thus, for example, we and others have found mental importance to emotion theory. A primary example can that marked increases in heart rate distinguish one subset of be found in the metaphors we use to describe our emotional basic emotions (anger, fear, sadness) from another subset that experiences (e.g., heat and pressure metaphors associated with shows little change in heart rate (enjoyment, surprise), and from anger). These metaphors appear to be based on the actual disgust, which shows heart rate slowing (Ekman, Levenson, & physiological changes that occur during those emotional states Friesen, 1983; Levenson, 1992). (Lakoff, 1987, 1993; Levenson, 2003; Marchitelli & Levenson, Those who believe that facial expressions and patterns of 1992). I have endorsed the neo-Jamesian view that our subjective autonomic nervous system activity are defining features of emotional experience is largely constructed from the interocep- basic emotions do not typically think that basic emotions are tive and proprioceptive information that becomes available differentiated by a single unique facial action or a single when emotions (and particularly basic emotions) produce unique physiological change, but rather that they are distin- changes in autonomic and somatic nervous system activity guished by unique configurations of multiple facial actions or (Levenson, 1999). For most people these physical sensations of multiple physiological responses. As a result, there are are not very precise (Pennebaker, 1982), but they are consistent sizeable literatures that have searched for these configura- enough to produce broad cross-cultural agreement in surveys of tions, and numerous spirited debates about the quality of the physical sensations that are associated with particular emo- existing evidence in support of the specificity of facial and tions (Wallbott & Scherer, 1988). Recently, we developed a autonomic configurations (Levenson, 1992; Zajonc & laboratory paradigm for assessing how closely subjective McIntosh, 1992). emotional experience tracks underlying physiological and 384 Emotion Review Vol. 3 No. 4

behavioral activity, finding this tracking is quite close during with what Zajonc characterized many years ago as requiring no emotional episodes (Mauss, Levenson, McCarter, Wilhelm, & prior cognition. Gross, 2005). Moreover, the tracking between subjective expe- Having said that, it must be noted that once this rapid first rience and physiology is particularly close in those who have level of processing occurs, there is usually a “secondary response” received extensive training in attending to and monitoring their in humans in which the original elicitor, the activated visceral activity (Sze, Gyurak, Yuan, & Levenson, 2010). response, the personal and social consequences of the activated response, and myriad other factors (social learning, cultural Question 8: What are the minimal cognitive prerequisites beliefs, values, past experiences, etc.) become involved. These for the occurrence of a basic emotion? factors can produce a secondary emotional response, which can be more of the same basic emotion, a different basic emotion, a This question recalls the lively exchanges between Zajonc and more complex emotion (e.g., a “self-conscious” emotion such Lazarus over the issue of whether cognition was necessary for as embarrassment, shame, pride, and guilt), or some combina- emotion (Lazarus, 1981, 1984; Zajonc, 1980, 1984). In the end, tion of emotions. I believe this secondary response is clearly this debate essentially came down to how “cognition” was “cognitive” in the ways that Lazarus imagined. defined. If cognition was defined as “anything the brain does,” Recently, we have been able to show in neurological patients then Lazarus’ position that all emotion requires cognition held instances in which the initial, simple emotional response sway. If cognition was defined as a more lengthy, considered remains intact but the more complex secondary emotional process of weighing possibilities, considering alternatives, and response is virtually eliminated (Sturm, Ascher, Miller, & planning responses, then Zajonc’s position that emotion could Levenson, 2008; Sturm, Rosen, Allison, Miller, & Levenson, occur without cognition won out. Today, this argument would 2006). This evidence for dissociation implies that the neural still largely resolve into a question of how you defined cogni- circuitry for the two kinds of responses is in fact different. tion and, to an extent, how you defined emotion. For example, reflexes such as the startle response are known to be mediated References at the level of the spinal cord (Davis, Gendelman, Tischler, & Adolphs, R. (1999). Social cognition and the human brain. Trends in Gendelman, 1982) and thus, it would be difficult to make the Cognitive Science, 3, 469–479. case for the involvement of cognition. However, many would Boucher, J. D. (1983). Antecedents to emotions across cultures. In S. H. Irvine & J. W. Berry (Eds.), Human assessment and cultural factors argue that the startle reflex is not an emotion (Ekman, Friesen, (pp. 407–420). New York, NY: Plenum. & Simons, 1985). Darwin, C. (1872). The expression of the emotions in man and animals. In the 25 years since the Lazarus–Zajonc debate, there has London, UK: Murray. been a remarkable explosion of knowledge about the neural Davis, M., Gendelman, D. S., Tischler, M. D., & Gendelman, P. M. (1982). circuitry of emotion derived from affective neuroscience. This A primary acoustic startle circuit: Lesion and stimulation studies. work has deeply increased our understanding of the automatic, Journal of Neuroscience, 2, 791–805. Cognition & Emotion minimally “cognitive” level of processing that Zajonc champi- Ekman, P. (1992). An argument for basic emotions. , 6, 169–200. oned. We now know that there is very early, low-level processing Ekman, P., & Friesen, W. V. (1969). The repertoire of nonverbal behavior: of incoming sensory information for its emotional relevance in Categories, origins, usage, and coding. Semiotica, 1, 49–98. brain centers such as the amygdala, that this processing is fast Ekman, P., & Friesen, W. V. (1971). Constants across cultures in the face and automatic, that it can occur absent of conscious aware- and emotion. Journal of Personality and Social Psychology, 17, 124–129. ness, and that it has the capacity to influence subsequent Ekman, P., & Friesen, W. V. (1975). Unmasking the face: A guide to recog- behaviors profoundly (Adolphs, 1999; LeDoux, 2000; Rosen nizing emotions from facial clues. Englewood Cliffs, NJ: Prentice Hall. Ekman, P., & Friesen, W. V. (1978). Facial action coding system. Palo Alto, & Levenson, 2009). CA: Consulting Psychologists Press. Thus, the cognitive prerequisites for the occurrence of a Ekman, P., Friesen, W. V., & Simons, R. C. (1985). Is the startle reaction an basic emotion are “minimal.” Primitive, phylogenetically emotion? Journal of Personality and Social Psychology, 49, 1416–1426. ancient brain centers like the amygdala are constantly assessing Ekman, P., & Heider, K. G. (1988). The universality of a contempt expression: incoming sensory information for patterns of input that are asso- A replication. Motivation and Emotion, 12, 303–308. ciated with species-specific challenges and opportunities. When Ekman, P., Levenson, R. W., & Friesen, W. V. (1983). Autonomic nervous Science 221 such a pattern is detected, it is matched to one of a limited set system activity distinguishes among emotions. , , 1208–1210. Elfenbein, H. A., Beaupre, M., Levesque, M., & Hess, U. (2007). Toward a of basic emotions, each of which has been shaped by evolution dialect theory: Cultural differences in the expression and recognition of to organize behavior and physiology in ways that provide a posed facial expressions. Emotion, 7, 131–146. generalized response that is most likely to deal successfully Fredrickson, B. L., & Levenson, R. W. (1998). Positive emotions speed with the eliciting situation. This all occurs rapidly and does not recovery from the cardiovascular sequelae of negative emotions. require any conscious awareness or higher order cognitive Cognition & Emotion, 12, 191–220. processing (e.g., planning, deliberation, measured consideration Friesen, W. V. (1972). Cultural differences in facial expressions in a social situation: An experimental test of the concept of display rules of response alternatives). The brain is, of course, highly . Unpublished doctoral dissertation, University of California, San involved in all of this, but the circuitry is quite ancient. Whether Francisco, CA. this is considered “cognition” or not is a question of definition. George, M. S., Wassermann, E. M., Williams, W. A., Steppel, J., But, I believe this level of emotion processing is quite consistent Pascual-Leone, A., Basser, P. … Post, R. M. (1996). Changes in mood Levenson Basic Emotion Questions 385

and hormone levels after rapid-rate transcranial magnetic stimulation Athens Olympic Games. Journal of Personality and Social Psychology, (rTMS) of the prefrontal cortex. Journal of Neuropsychiatry and 91, 568–581. Clinical Neurosciences, 8, 172–180. Matsumoto, D., & Willingham, B. (2009). Spontaneous facial expressions Gray, J. M., Young, A. W., Barker, W. A., Curtis, A., & Gibson, D. (1997). of emotion of congenitally and noncongenitally blind individuals. Impaired recognition of disgust in Huntington’s disease gene carriers. Journal of Personality and Social Psychology, 96, 1–10. Brain, 120, 2029–2038. Mauss, I. B., Levenson, R. W., McCarter, L., Wilhelm, F. H., & Gross, J. J. Gross, J. J., & Levenson, R. W. (1997). Hiding feelings: The acute effects (2005). The tie that binds? Coherence among emotion experience, of inhibiting negative and positive emotion. Journal of Abnormal behavior, and physiology. Emotion, 5, 175–190. Psychology, 106, 95–103. Olney, N. T., Goodkind, M., Lomen-Hoerth, C., Whalen, P., Williamson, C., Hochschild, A. R. (1979). Emotion work, feeling rules, and social structure. Holly, D. … Rosen, H. J. (2011). Behavior, physiology, and experi- American Journal of Sociology, 84, 551–575. ence of pseudobulbar affect in patients with ALS. Unpublished Izard, C. E. (1971). The face of emotion. New York, NY: Appleton- manuscript. Century-Crofts. Pennebaker, J. W. (1982). The psychology of physical symptoms. New York, Keltner, D. (1995). Signs of appeasement: Evidence for the distinct displays NY: Springer-Verlag. of embarrassment, amusement, and shame. Journal of Personality and Reis, D. J., & Gunne, L. M. (1965). Brain catecholamines: Relation to the Social Psychology, 68, 441–454. defense reaction evoked by amygdaloid stimulation in cat. Science, 149, Keltner, D., & Buswell, B. N. (1997). Embarrassment: Its distinct form and 450–451. appeasement functions. Psychological Bulletin, 122, 250–270. Reis, D. J., Miura, M., Weinbren, M., & Gunne, L. M. (1967). Brain cate- Lakoff, G. (1987). Women, fire, and dangerous things. Chicago, IL: cholamines: Relation to defense reaction evoked by acute brainstem University of Chicago Press. transection in cat. Science, 156, 1768–1770. Lakoff, G. (1993). The contemporary theory of metaphor. In A. Ortony Rosch, E., & Lloyd, B. B. (1978). Principles of categorization. In E. Rosch (Ed.), Metaphor and thought (pp. 202–251). New York, NY: Cambridge (Ed.), Cognition and categorization (pp. 28–49). Hillsdale, NJ: Lawrence University Press. Erlbaum. Lawrence, A. D., Calder, A. J., McGowan, S. W., & Grasby, P. M. (2002). Rosch, E., & Mervis, C. B. (1975). Family resemblances: Studies in the Selective disruption of the recognition of facial expressions of anger. internal structures of categories. Cognitive Psychology, 7, 573–605. Neuroreport, 13, 881–884. Rosen, H. J., & Levenson, R. W. (2009). The emotional brain: Combining Lazarus, R. S. (1981). A cognitivist’s reply to Zajonc on emotion and insights from patients and basic science. Neurocase, 15, 173–181. cognition. American Psychologist, 36, 222–223. Russell, J. (1991). In defense of a prototype approach to emotion concepts. Lazarus, R. S. (1984). On the primacy of cognition. American Psychologist, Journal of Personality and Social Psychology, 60, 37–47. 39, 124–129. Scherer, K. R., Summerfield, A. B., & Wallbott, H. G. (1983). Cross- LeDoux, J. E. (2000). Emotion circuits in the brain. Annual Review of national research on antecedents and components of emotion: Neuroscience, 23, 155–184. A progress report. Social Science Information, 22, 355–385. Levenson, R. W. (1988). Emotion and the autonomic nervous system: Soto, J. A., Levenson, R. W., & Ebling, R. (2005). Cultures of moderation A prospectus for research on autonomic specificity. In H. L. Wagner and expression: Emotional experience, behavior, and physiology in (Ed.), Social psychophysiology and emotion: Theory and clinical Chinese Americans and Mexican Americans. Emotion, 5, 154–165. applications (pp. 17–42). Chichester, UK: John Wiley & Sons. Sprengelmeyer, R., Young, A. W., Calder, A. J., & Karnat, A. (1996). Loss Levenson, R. W. (1992). Autonomic nervous system differences among of disgust: Perception of faces and emotions in Huntington’s disease. emotions. Psychological Science, 3, 23–27. Brain, 119, 1647–1665. Levenson, R. W. (1994a). Human emotion: A functional view. In P. Ekman Sprengelmeyer, R., Young, A. W., Sprenglemeyer, A., Calder, A. J., & R. J. Davidson (Eds.), The nature of emotion: Fundamental questions Rowland, D., Perrett, D., … Lange, H. (1997). Recognition of facial (pp. 123–126). New York, NY: Oxford. expressions: Selective impairment of specific emotions in Huntington’s Levenson, R. W. (1994b). The search for autonomic specificity. In P. Ekman disease. Cognitive Neuropsychology, 14, 839–879. & R. J. Davidson (Eds.), The nature of emotion: Fundamental questions Sturm, V. E., Ascher, E. A., Miller, B. L., & Levenson, R. W. (2008). (pp. 252–257). New York, NY: Oxford. Diminished self-conscious emotional responding in frontotemporal Levenson, R. W. (1999). The intrapersonal functions of emotion. Cognition lobar degeneration patients. Emotion, 8, 861–869. & Emotion, 13, 481–504. Sturm, V. E., Rosen, H. J., Allison, S., Miller, B. L., & Levenson, R. W. Levenson, R. W. (2003). Blood, sweat, and fears: The autonomic architec- (2006). Self-conscious emotion deficits in frontotemporal lobar ture of emotion. In P. Ekman, J. J. Campos, R. J. Davidson, & F. B. M. degeneration. Brain, 129(Pt. 9), 2508–2516. de Waal (Eds.), Emotions inside out: 130 years after Darwin’s The Susskind, J. M., & Anderson, A. K. (2008). Facial expression form and expression of the emotions in man and animals (pp. 348–366). New function. Communicative and Integrative Biology, 1, 148–149. York, NY: The New York Academy of Sciences. Sze, J. A., Gyurak, A., Yuan, J. W., & Levenson, R. W. (2010). Coherence Levenson, R. W., Ekman, P., Heider, K., & Friesen, W. V. (1992). Emotion between emotional experience and physiology: Does body awareness and autonomic nervous system activity in the Minangkabau of West training have an impact? Emotion, 10, 803–814. Sumatra. Journal of Personality & Social Psychology, 62, 972–988. Tomkins, S. S. (1962). Affect, imagery, consciousness. Volume 1. The positive Levenson, R. W., Soto, J., & Pole, N. (2007). Emotion, biology, and culture. affects. New York, NY: Springer. In S. Kitayama & D. Cohen (Eds.), Handbook of cultural psychology Tooby, J., & Cosmides, L. (1990). The past explains the present: Emotional (pp. 780–796). New York, NY: Guilford Press. adaptations and the structure of ancestral environments. Ethology & Marchitelli, L., & Levenson, R. W. (1992, October). When couples con- Sociobiology, 11, 375–424. verse: The language and physiology of emotion. Paper presented at the Tracy, J. L., & Matsumoto, D. (2008). The spontaneous expression Society for Psychophysiological Research, San Diego, CA. of pride and shame: Evidence for biologically innate nonverbal dis- Matsumoto, D., Kudoh, T., Scherer, K., & Wallbott, H. (1988). Antecedents plays. Proceedings of the National Academy of Sciences USA, 105, of and reactions to emotions in the United States and Japan. Journal of 11655–11660. Cross-Cultural Psychology, 19, 267–286. Tracy, J. L., & Robins, R. W. (2008). The nonverbal expression of pride: Matsumoto, D., & Willingham, B. (2006). The thrill of victory and the Evidence for cross-cultural recognition. Journal of Personality and agony of defeat: Spontaneous expressions of medal winners of the 2004 Social Psychology, 94, 516–530. 386 Emotion Review Vol. 3 No. 4

Tsai, J. L., Knutson, B., & Fung, H. H. (2006). Cultural variation in Yuan, J. W., McCarthy, M., Holley, S. R., & Levenson, R. W. (2010). affect valuation. Journal of Personality and Social Psychology, 90, Physiological down-regulation and positive emotion in marital interaction. 288–307. Emotion, 10, 467–474. Wallbott, H. G., & Scherer, K. R. (1988). How universal and specific is Zajonc, R. B. (1980). Feeling and thinking: Preferences need no inferences. emotional experience?: Evidence from 27 countries on five continents. American Psychologist, 35, 151–175. In K. R. Scherer (Ed.), Facets of emotion: Recent research (pp. 31–56). Zajonc, R. B. (1984). On the primacy of affect. American Psychologist, 39, Hillsdale, NJ: Lawrence Erlbaum. 117–123. Wilson, S. A. K. (1924). Some problems in neurology, II: Pathological Zajonc, R. B., & McIntosh, D. N. (1992). Emotions research: Some promis- laughing and crying. Journal of Neurology and Psychopathology, 4, ing questions and some questionable promises. Psychological Science, 299–333. 3, 70–74.